Project-based learning is among the most common learning approaches aimed at conveying professional standards and best practices to students in programming courses. However, team projects commonly impose problems related to responsibility sharing, such as low effort or inequality in contributions. This paper presents a collaborative programming assessment system featuring a code quality assessment function with specific metrics to measure individual contributions. Student engagement data is used to detect nonconformities in collaboration using a learning analytical approach. Latent profile analysis was used to detect four theoretical team profiles differentiated by team effort (2 levels) and within-team collaboration (2 levels). We demonstrated the efficacy of assessing code to evaluate team dynamics and student behaviour, wherein efforts to resolve coding style failures can be used as a proxy by which to estimate the taskwork awareness of team members. Submission data from 146 students in 41 web-programming projects revealed four behavioural patterns that could potentially hinder the effective functioning of programming teams: free-riding, social loafing, the bystander effect, and lone wolves. We also demonstrated the applicability of automated programming assessment systems to the monitoring of learning progress, thereby facilitating timely interventions to correct difficulties at the team level.

Ontology is an effective formal representation of knowledge used commonly in artificial intelligence, semantic web, software engineering and information retrieval. Typically, ontologies are constructed by domain experts using domain knowledge and domain documents. However, manual acquisition of ontologies from domain documents consumes high costs. We present a support system for Vietnamese ontology construction using pattern-based mechanisms of discovering Vietnamese concepts and conceptual relations from Vietnamese text documents. As there are very few existing taxonomies constructed in Vietnamese, we use non-taxonomy based approach. The combination of association rule mining and lexical pattern based learning was used as our major method of concept extraction and conceptual relation detection. From the experiments, we show that this is a feasible solution.